(103 days)
The NMP22 BladderChek Test is indicated for professional and prescription home use as an aid in monitoring bladder cancer patients, in conjunction with standard diagnostic procedures.
The NMP22 BladderChek Test for nuclear matrix protein NMP22 is an immunochromatographic assay utilizing monoclonal antibodies in a lateral flow strip encased in plastic. Two different antibodies are used, one as a capture and one as a reporter antibody. Unprocessed patient urine is added to the sample well of the cartridge and allowed to react with the colloidal gold conjugated reporter antibody. If the antigen is present in urine, it will interact with the reporter conjugate to form an immune complex. The reaction mixture flows through the membrane, which contains zones of immobilized antibodies. In the Test (T) zone, antigen-conjugate complexes are trapped by the capture antibody, forming a visible line if the concentration of antigen in urine is elevated. The procedural Control (C) zone contains an immobilized goat anti-mouse IgG-specific antibody that will capture the conjugated antibody independently of the presence or absence of the antigen, thereby always producing a visible line in the Control window. This procedural control assures the operator that each device is working properly.
Here's a summary of the acceptance criteria and study details for the NMP22 BladderChek™ Kit, based on the provided text:
1. Acceptance Criteria and Reported Device Performance
Although explicit "acceptance criteria" (numerical thresholds that must be met for approval) are not directly stated in the summary, the overall performance goals are implied by successful demonstration of substantial equivalence to the predicate device and satisfactory clinical performance. The table below presents the clinical performance metrics reported for the NMP22 BladderChek™ Test and, for comparison, the predicate device.
| Performance Metric | Acceptance Criteria (Implied) | Reported Device Performance (NMP22 BladderChek™) | Predicate Device Performance (BTA Stat) |
|---|---|---|---|
| Clinical Sensitivity | Comparable or better than predicate | 45.9% (95% CI: 35.8%, 56.3%) (45/98) | 67% (95% CI: 60%, 73%) |
| Clinical Specificity | Comparable or better than predicate | 86.3% (95% CI: 83.2%, 89.0%) (492/570) | 70% (95% CI: 61%, 79%) |
| Positive Predictive Value (PPV) | Comparable or better than predicate | 36.6% (95% CI: 28.1%, 45.8%) at 14.7% incidence | 19.8% to 48.8% (theoretical range due to retrospective data) |
| Negative Predictive Value (NPV) | Comparable or better than predicate | 90.3% (95% CI: 87.5%, 92.6%) at 14.7% incidence | (Not calculated for predicate due to retrospective data) |
| Overall Accuracy | Comparable or better than predicate | 80.3% | 67.9% |
| Reproducibility (Lab Techs) | High concordance across lots, users, and days | 99.2% overall correct reads (1786/1800) | "Nearly total agreement" (near cut-off exceptions) |
| Reproducibility (Lay vs. Professional) | High concordance | 100% concordance (lay vs. professional) on precision panel | N/A |
| Concordance (Lay vs. Professional – Real-world) | High agreement | 96.4% (professional re-read lay user device); 95.6% (professional re-test lay user urine) | N/A |
| Concordance with Microplate Test | Good concordance | 91.2% overall (198/217) | N/A |
Summary of the Study Proving Acceptance Criteria:
The device's performance was evaluated through non-clinical and clinical studies to demonstrate substantial equivalence to the predicate device, BTA Stat.
2. Sample Size Used for the Test Set and Data Provenance:
-
Clinical Test Set:
- Sample Size: 668 patients.
- Data Provenance: Prospective clinical trial conducted at 23 sites. The country of origin is not explicitly stated but implied to be the US given the submission to the FDA.
-
Non-Clinical Test Set (Reproducibility by Laboratory Technicians):
- Sample Size: 150 individual reads per panel level (3 readers x 10 devices per panel x 5 days) for 4 NMP22 levels (0, 5, 15, 25 U/mL). Total 600 reads per lot. Experiment conducted on three separate lots, so 1800 total unique device reads.
- Data Provenance: Not specified, but likely internal lab studies.
-
Non-Clinical Test Set (Reproducibility by Lay Users Compared to Professional Readers):
- Sample Size: 5 lay readers (10 results per level) and 2 professional readers (10 results per level) for a three-level precision panel (2, 10, 15 U/mL). In total, 50 lay user results and 20 professional reader results.
- Data Provenance: Not specified, but likely internal lab studies.
-
Non-Clinical Test Set (Performance of Lay Users Compared to Professionals – Field Studies):
- Sample Size: 137 volunteers (aged 50+), across 3 sites.
- Data Provenance: Field studies at three unspecified locations.
-
Non-Clinical Test Set (Concordance of NMP22 BladderChek and NMP22 Test Kit (microplate)):
- Sample Size: 217 voided urine samples.
- Data Provenance: Urology clinics.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications:
-
Clinical Test Set:
- Number of Experts: Not explicitly stated, but the ground truth for bladder cancer status was determined by physicians conducting cystoscopies and subsequent pathological examination of resected lesions. These are highly qualified medical professionals (e.g., urologists and pathologists).
- Qualifications: "Physicians conducting the cystoscopies" and "pathologically determined to be malignant."
-
Non-Clinical Test Set (Reproducibility):
- Laboratory Technicians: 3 laboratory technicians. Qualifications not specified, but implied to be trained in laboratory procedures.
- Professional Readers (Lay User Reproducibility): 2 professional readers. Qualifications not specified.
4. Adjudication Method for the Test Set:
-
Clinical Test Set: The primary ground truth was established by cystoscopy findings, confirmed by pathology for resected lesions. Physicians were blinded to the device results, suggesting an independent assessment of ground truth. There is no mention of a specific adjudication method (like 2+1 or 3+1) among multiple experts for establishing the cancer status itself, rather it relies on standard medical diagnostic procedures.
-
Non-Clinical Test Set (Reproducibility):
- For the lab technician reproducibility, the ground truth was based on the known NMP22 concentration of the prepared specimen panels (0, 5, 15, 25 U/mL).
- For lay user vs. professional reproducibility, the ground truth was based on the known NMP22 concentration of the prepared samples (2, 10, 15 U/mL). Concordance was measured between readers rather than against an external adjudicator.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:
- No specific MRMC comparative effectiveness study was mentioned comparing human readers with and without AI assistance (as the device is a diagnostic kit, not an AI).
- However, the study did involve a comparison of lay users vs. professional readers using the device, which is a form of multi-reader evaluation. There was 100% concordance between lay and professional readers on precision panels and high concordance (95.6% - 96.4%) in field studies for real urine samples, indicating that the device design allows for consistent interpretation across different user types.
6. Standalone Performance:
- Yes, a standalone performance study was done. The entire clinical study assessed the performance of the NMP22 BladderChek™ Test alone (algorithm/device only, without human-in-the-loop assistance influencing the result) against the cystoscopy/pathology ground truth. The reported sensitivity, specificity, PPV, and NPV are all measures of this standalone performance.
7. Type of Ground Truth Used:
-
Clinical Test Set: The ground truth was based on cystoscopy findings, with confirmation by pathology for malignant lesions. "Patients were considered negative if no tumor was seen endoscopically, or, if a lesion was seen, was pathologically determined to be nonmalignant. Patients were considered positive for bladder cancer if a tumor was seen during cystoscopy, and, if removed was pathologically determined to be malignant." This is a combination of observational medical findings and definitive pathological diagnosis.
-
Non-Clinical Test Set (Reproducibility): The ground truth was based on known concentrations of NMP22 in prepared urine specimens.
8. Sample Size for the Training Set:
- Not explicitly stated. The submission describes validation studies but does not detail a separate training set size for the development of the device itself (e.g., for antibody selection or cut-off determination). This is typical for a diagnostic kit where the "training" (e.g., cut-off determination) often happens during product development and optimization, rather than a distinct "training set" in the context of an AI algorithm.
9. How the Ground Truth for the Training Set Was Established:
- Not explicitly detailed as a distinct training set is not explicitly mentioned. For the development and optimization of such a diagnostic device (e.g., determining the NMP22 threshold for a positive result like 10 U/mL mentioned in the microplate comparison), ground truth would typically be established similarly to the clinical test set: using urine samples from known bladder cancer patients and healthy controls, confirmed by established diagnostic methods like cystoscopy and pathology. The 10 U/mL cutoff for the NMP22 microplate test is stated, implying that this was an established threshold for a positive result.
§ 866.6010 Tumor-associated antigen immunological test system.
(a)
Identification. A tumor-associated antigen immunological test system is a device that consists of reagents used to qualitatively or quantitatively measure, by immunochemical techniques, tumor-associated antigens in serum, plasma, urine, or other body fluids. This device is intended as an aid in monitoring patients for disease progress or response to therapy or for the detection of recurrent or residual disease.(b)
Classification. Class II (special controls). Tumor markers must comply with the following special controls: (1) A guidance document entitled “Guidance Document for the Submission of Tumor Associated Antigen Premarket Notifications (510(k)s) to FDA,” and (2) voluntary assay performance standards issued by the National Committee on Clinical Laboratory Standards.